Introduction

Idiopathic scoliosis (IS) is a three-dimensional spinal deformity characterized by lateral deviation and rotation of the spine, often associated with a reduction in its normal kyphotic curvature [1, 2]. Biomechanical and morphological variations are prevalent among individuals with IS [3]. They encompass discrepancies in limb–trunk asymmetry, postural malalignment, compensatory motion patterns, alterations in rib, pelvis, arm, and skull morphology, as well as reduction in the overall body stability/mobility. These changes result in postural deformity of the spine with chronic repetitive asymmetric stress on it [4]. Eventually, various problems such as muscle imbalance, functional limitation, low back pain, and their negative effects on the psychosocial well-being and body image might ensure [5].

IS has also been identified as a potential risk factor for the development of chronic low back pain (CLBP) in adolescents [6, 7]. The latter was reported with prevalence rates ranging from 27.5% to 72% in adolescents with IS [8, 9] having negative impact on physical and psychological functions as well as quality of life [10]. The etiology of CLBP includes asymmetric facet joint loading (resulting in arthritis/synovitis), muscular pain due to eccentric load on the curvature apex, and intervertebral disk problems [11, 12]. However, the relationship between CLBP and IS has yet not been well established [13].

The thoracolumbar fascia (TLF) is an anatomical structure composed of dense connective tissue layers in the lumbar region. It comprises aponeurotic and fascial planes united around the spinal muscles. The TLF plays an essential role in postural alignment and stability/movement of the spinal column [14, 15]. There is growing evidence as regards the role of TLF in individuals with CLBP [16]. For instance, it has been found to be thicker in subject with CLBP [17]. Moreover, any pathology in the TLF can impair its flexibility and elastic characteristics, thereby restricting the trunk range of motion [18]. Herewith, the TLF thickness has not been examined in individuals with IS. To this end, the hypothesis of our study was that there could be a relationship between CLBP and TLF in individuals with IS, whereby the thickness might affect trunk motion and flexibility. Accordingly, the present study aimed to evaluate the TLF and its relationship with CLBP in IS.

Materials and Methods

Study design

The data were collected between July–October 2023. The participants were recruited from individuals who were referred to the Department of Physical Therapy and Rehabilitation at university hospital. All individuals were informed prior to inclusion in the study, and written informed consent was collected from participants aged 18 and above. For participants under the age of 18, informed consent was also obtained from the parents.

Participants

A total of 60 individuals with IS were enrolled and divided into three groups: The painful group consisted of individuals who had pain for more than three months, with a severity score of 3 or higher based on the visual analog scale (VAS); the non-painful group comprised individuals with no pain; and the healthy group had no spine pathology or low back pain.

Individuals with IS (above the age of 10) were included if they had a left lumbar curve with a Cobb angle of more than 15 degrees. Healthy individuals (above the age of 10) were recruited if they had no back pain (VAS score = 0), orthopedic, neurological, or rheumatic problems that could affect spinal health. Individuals with a history of lower extremity or spinal region surgery/trauma were excluded.

Outcome measures

Age, gender, height, weight, and body mass index were collected from all participants. Cobb angle and axial trunk rotation were noted in the scoliosis group.

Pain severity was evaluated using the Turkish version of the short form of the McGill Pain Questionnaire (SF-MPQ) to assess the sensory/affective characteristics of pain and its intensity [19]. The primary component comprises 15 descriptors (11 sensory; four affective) that are assessed on an intensity scale, with ratings ranging from 0 (none) to 3 (severe). The second section includes five words designed to determine the intensity of pain. The third section entails the participant’s evaluation of the current pain intensity using VAS. Three pain scores are computed based on the sum of the intensity rank values assigned to the chosen words for sensory, affective, and overall descriptors.

Thickness of the TLF was measured at the L3 lumbar vertebra level by ultrasonography and using a previously described standardized protocol. Bilateral measurements (mm) were taken as subjects lay relaxed in prone position, and the transducer was localized vertical to the spine, approximately 2–3 cm lateral to the L3 spinous process. In order to mitigate the potential impact of thickness variations, three equidistant-spaced points were assessed per image for TLF and averaged for analysis (Fig. 1) [20]. A single physiatrist with 25-year experience in musculoskeletal ultrasound performed all the measurements.

Fig. 1
figure 1

Ultrasound imaging (short-axis view) over the left (A) and right (B) paravertebral areas. Red rectangles indicate the measurement sites of the thoracolumbar fascia

Full size image

Active trunk range of motions were assessed during flexion/extension, rotation, and lateral flexion using the universal goniometer. Prior to each evaluation, individuals were provided with an explanation of the motion and given an opportunity to perform a trial. During the assessment of trunk flexion and extension, the therapist stood on the lateral side of the individual. The goniometer’s pivot rested on the hip at the lateral projection of the lumbosacral joint. The stable arm was perpendicular to the ground, while the moving arm traced the lateral projection of the trunk toward the axilla. Special attention was given to preventing any hip joint motion throughout the measurement. For lateral flexion measurement, the therapist stood behind the individual. The goniometer’s pivot was at the midpoint of the lumbosacral joint, with the stable arm parallel to the posterior superior iliac spine and the ground. The moving arm followed the spinal processes of the lumbar vertebrae toward the cranial side. The therapist ensured no rotation, flexion, or extension of the trunk during the measurements. During the assessment for lumbar rotation, the therapist stood behind the seated individual. The pivot point was centered on the head, with the fixed goniometer arm parallel to the ground. The moving arm followed the acromion in the opposite direction of rotation. In both right and left rotation measurements, careful attention was paid to prevent any cranial rotation by the subject [21, 22].

Flexibility of the trunk was evaluated using the sit-and-reach test [23]. A tape measure was placed on a table at the soles of the feet. The individuals leaned forward slowly reaching as far as possible while keeping their hands adjacent to each other. After a familiarization test, the best of three trials was recorded (cm).

Statistical analysis

Data were analyzed using IBM SPSS 27 for Windows software. Descriptive statistical methods were initially employed. Normal distribution was assessed using the Kolmogorov–Smirnov test. To assess differences among the demographic characteristics of the groups, chi-square test was employed for gender, ANOVA test for age, height, and body weight, and Mann–Whitney U test for Cobb angle and rotation. When comparing trunk motions between groups, ANOVA and Kruskal–Wallis tests were used, where appropriate. Correlation analyses were performed using Pearson and Spearman correlation coefficients, as appropriate. To determine significant differences among groups, Bonferroni correction was employed for post-hoc analysis.

In assessing intra-rater reliability, we employed the intra-class correlation coefficient (ICC; two-way mixed effect model (absolute agreement)) based on three repeated ultrasound measurements. The interpretation of ICC values followed established criteria: poor (< 0.40), fair (0.40–0.59), good (0.60–0.74), and excellent (0.75–1.00) [24]. Additionally, we computed Cronbach’s alpha, Standard Error of Measurement (SEM), and minimum detectable changes (MDC = √2 × 1.96 × SEM) in accordance with the recommendations of Bland and Altman [25].

For determining an adequate sample size, a power analysis was conducted using G*Power Version 3.1.9.2 software. The effect size (f) obtained from a pilot study with 30 participants on scoliotic and healthy individuals was found to be 0.56. With a 95% power and type I error rate of 5% (α = 0.05), the study aimed to include 54 participants. However, considering equal distribution among groups and accounting for environmental and unforeseen factors, 60 individuals who voluntarily agreed to participate were enrolled.

Results

Demographics of the participants are presented in Table 1. No significant differences were observed among the groups in terms of age, height, body weight, and body mass index (all p > 0.05). There was no significant difference in Cobb angle and axial trunk rotation degrees between the non-painful and painful groups (both p > 0.05).

Table 1 Demographic characteristics of individuals according to groups
Full size table

Differences in right and left TLF thicknesses were observed among the groups (Table 2). Post-hoc tests revealed that the right TLF thickness of the painful group was greater than the non-painful (p = 0.007) and the healthy (p < 0.001) groups. The right TLF thickness of the non-painful group was also greater than that of the healthy group (p = 0.015). The left TLF thickness of both the non-painful (p = 0.005) and painful (p = 0.001) groups was greater than that of the healthy group, whereby the scoliosis groups were similar in this aspect (p > 0.05). Regarding trunk motions, left lateral flexion was decreased in the painful (vs. healthy) group (p = 0.025).

Table 2 Comparison of thoracolumbar fascia thickness and trunk movements between groups
Full size table

With respect to the relationship between TLF thickness, trunk motion, flexibility, and pain (Table 3), no correlation was found in either the non-painful or the healthy group (all p > 0.05). In the painful group, moderate and negative correlations were found between the right TLF thickness and trunk flexion (r = −0.540, p = 0.01), trunk extension (r = −0.514, p = 0.02), and left rotation (r = −0.499, p = 0.02). There was also a moderate negative correlation between the left TLF thickness and trunk extension (r = −0.528, p = 0.02) and a moderate positive correlation between the right TLF thickness and MPQ (r = 0.562, p = 0.01).

Table 3 Relationship among thoracolumbar fascia thickness, trunk motion, flexibility, and pain
Full size table

Intra-rater reliability

The ultrasound outcome measurements exhibited consistently excellent reliability across three repeated assessments. This evaluation was based on key statistical indicators, including Cronbach’s α, Intra-class Correlation Coefficient (ICC), Standard Error of Measurement (SEM), and Minimal Detectable Change (MDC), all thoroughly summarized in Table 4.

Table 4 Intra-rater reliability of the ultrasound outcome measurements
Full size table

Discussion

This study revealed that the TLF thicknesses were greater on either side of the curve in individuals with left-sided lumbar scoliosis (vs. healthy peers). Additionally, between individuals with IS, those who had CLBP had greater fascia thickness on the concave side when compared to non-painful individuals. Although TLF thickness was found to be positively correlated with pain, there was no association between fascial thickness and trunk motion in healthy or non-painful IS subjects. However, TLF thickness was particularly observed to be related to trunk motion limitation in the sagittal and horizontal planes in IS with CLBP.

Studies in the literature have confirmed asymmetries of the paravertebral and abdominal wall muscles in individuals with IS. Using ultrasound measurements, Karina et al. [26] found significantly greater muscle thickness on the concave side of patients with mild IS (vs. controls). An observational study with magnetic resonance imaging and histology showed that muscle degeneration was more common on the concave side in patients with degenerative lumbar scoliosis [27]. However, studies focused on muscles and did not examine any TLF abnormality. In our study we found that in individuals with IS, fascia thickness was significantly different from healthy individuals on both the convex and concave sides. It is noteworthy that the fascia thickens in individuals with IS. We speculate that the TLF might have undergone remodeling over time in response to repetitive stress due to compensatory patterns of motion, postural deviation, or asymmetric loading.

Another ultrasonographic study conducted in individuals with CLBP lasting more than 12 months reported increased thickness and echogenicity of the perimuscular connective tissues forming the TLF [17]. Similarly, Pirri et al. [20] and Gumruk et al. [28] showed that the TLF was thicker in patients with non-specific CLBP, whereby the latter also found positive correlation between TLF thickness and pain intensity. Our results yielded greater concave TLF thickness in painful (vs. non-painful) IS and positive correlation between TLF thickness and pain on the concave side. At this point, although a causal relationship between thicker TLF and CLBP cannot be ascertained, the former can be considered as a predisposing factor—also in light of the pertinent literature.

In previous cross-sectional (ultrasonographic) studies, thicker TLF with decreased mobility was identified in individuals with non-specific CLBP vs. non-painful peers [17, 20, 29, 30]. The potential role of fascia in CLBP was proposed by Langevin et al. [31] where pain-related fear was thought to induce a cycle of decreased motion. In turn, it was proposed that an altered motion pattern may result in connective tissue remodeling of TLF—leading to inflammation, nervous tissue sensitization, and further decreased mobility. Moreover, changes in the properties of fascia, such as thickness, may restrict muscular extensibility and range of motion [32]. A follow-up investigation found that TLF shear strain during passive trunk flexion was reduced in people with CLBP by 56% [29]. These changes alter/limit daily motion, determining the flaccidity of the spine and modifying the pattern of motion [33]. According to our results, in the painful group, a negative correlation was observed between the thickness of the concave TLF and trunk flexion–extension and left rotation. Structural and functional abnormalities of the TLF could have contributed to motion restriction, which could also/actually be worsened by pain.

Main strength of the current study is the inclusion of individuals with IS of the single lumbar curve pattern. This enabled evaluations to be conducted specifically for a certain type of curvature. In both painful and non-painful groups, individuals were found to have similar curve magnitudes and axial rotation degrees. Therefore, the potential impact of curve magnitude and rotation degree on the results has been prevented. Additionally, the use of ultrasound provided reliable/objective morphological assessment of the TLF thickness. On the other hand, this study has some limitations, yet, the results cannot be generalized to individuals who have other curve types or magnitudes. Further, the fact that TLF thickness measurements were taken at a single (L3 vertebra) level could be another drawback. Nonetheless, our study results can add to the growing evidence of data as regards the association between TLF and CLBP in subjects with IS.

Conclusion

This study showed that when comparing individuals with IS to healthy counterparts, the thickness of the TLF exhibits a notable increase. Additionally, in individuals with CLBP, the fascia in the concave region is found to be thicker than that of non-painful peers. TLF thickness is also associated with limitation of trunk motion in the sagittal/horizontal planes in painful individuals. Last but not least, to our best notice, our study is the first to have investigated the TLF in individuals with IS and examined its relationship with pain. Future research should investigate the cause and effect relationship of the aforementioned findings in a wider group of IS patients with different disease features.